PREVENT Remote Sensing Practical Session (UOWM)

Chapters Template - PREVENT Project

Remote Sensing Practical Session (Exercises)

Start

Introduction

The following practical exercises are designed to integrate remote sensing technologies into disaster prevention and mitigation strategies, aligning with the core objectives of the PREVENT project. Each exercise focuses on real-world applications of geographic information systems (GIS) and satellite imagery analysis, equipping students with technical skills in spatial analysis, environmental monitoring, and decision-making for disaster management.

Index

Exercise 3

Objectives

Exercise 4

Exercise 1

Exercise 2

Index

Exercise 3

Objectives

Exercise 4

Exercise 1

Exercise 2

Objectives

These exercises serve to establish a conduit between academic research and professional practice, cultivating critical problem-solving skills and technical competencies that are imperative for contemporary disaster risk assessment and emergency response planning. The practical nature of this course is designed to equip students with the skills and knowledge necessary to enhance their employability and contribute to the resilience of their communities through the implementation of innovative, technology-driven solutions.

PREVENT - Prevention of natural disasters using deep technology for advanced HEI curricula Remote Sensing Practical Session (Exercises)

By engaging with satellite data from the Sentinel-1 and Sentinel-2 missions, along with Digital Elevation Models (DEM) and Geographic Information System (GIS) software, students will acquire practical experience in the detection and evaluation of environmental hazards, including but not limited to wildfires, floods, and landslides.

Exercise 1:

Students will analyze the impacts of wildfires using vegetation indices (NDVI) to detect areas that have been burned and assess the affected ecosystems. This analysis supports the PREVENT goal of assessing and responding to the risk ofwildfires.

Sentinel-2 data processing, NDVI computation, GIS mapping, burned area classification

link

Exercise 2:

The objective of this exercise is to provide students with the necessary skills to process SAR (synthetic aperture radar) data for the purpose of flood detection. This initiative is in alignment with PREVENT's overarching objective of enhancing flood monitoring through the utilization of remote sensing technologies.

SAR data interpretation, GIS thresholding for water detection, flood extent mapping.

link

Exercise 3:

Students receive instruction in the evaluation of landslide-prone areas through the implementation of topographic analysis and vegetation indices, thereby contributing to the objectives of PREVENT, namely the provision of early warning systems and the prediction of hazards.

DEM processing, slope/aspect analysis, NDVI based vegetation health assessment.

link

Exercise 4:

The utilization of Geographic Information System (GIS) spatial analysis by students is intended to ascertain the most suitable locations for the placement of forest observation towers. This approach is expected to contribute to the enhancement of fire prevention and early detection efforts, thereby providing direct support to the objectives of PREVENT.

Terrain visibility analysis, buffer zone application,site suitability modeling.

link

01

Exercise 1: Identifying Burned Areas Using NDVI (Sentinel-2 Data)

01

Objective

This exercise introduces students to satellite-based wildfire assessment using the Normalized Difference Vegetation Index (NDVI). By analyzing Sentinel-2 imagery, students will detect burned vegetation, quantify damage, and visualize affected areas. This process helps in post-fire monitoring and recovery planning, directly supporting PREVENT’s goal of integrating remote sensing for disaster prevention. NDVI is a numerical index used to measure the health of vegetation in an area. It is based on the reflections of radiation in the red (Red) and near-infrared (NIR) wavelengths:

• Red (B04): Vegetation absorption
• NIR (B08): Vegetation reflectance

NDVI is calculated using the following formula: NDVI=(NIR-RED)(NIR+RED)NDVI=(NIR+RED)(NIR-RED) NDVI values close to 1 indicate healthy vegetation. NDVI values close to 0 or negative indicate a lack of vegetation (e.g., burned areas, bare ground).

Expected Outcomes

- Understand how NDVI can be used to detect burned vegetation. - Gain hands-on experience in Sentinel-2 data processing. - Learn thresholding techniques for disaster mapping. - Develop GIS-based analysis skills for wildfire risk assessment.

Required Tools

Sentinel-2 data

QGIS

You can download required data from the link below

You can download the application from the link below

Steps to Complete Exercise 1

Step 1

Step 3

Step 2

Step 4

Step 5

Step 7

Step 6

02

Exercise 2: Flood Analysis Using Sentinel-1 SAR Data

02

Objective

This exercise introduces students to Synthetic Aperture Radar (SAR) analysis for detecting and analyzing flooded areas. Using Sentinel-1 SAR data, students will identify regions affected by flooding, measure the extent of the inundation, and generate a flood hazard map. The ability to detect floods despite cloud cover and nighttime conditions makes SAR technology a critical tool in disaster management, aligning with the PREVENT project’s goal of integrating deep technology for early warning systems. What is SAR data? SAR data relies on radar to record reflections from the Earth's surface. Surfaces covered with water exhibit low reflectivity, while other surfaces (such as soil or vegetation) exhibit high reflectivity.

Expected Outcomes

- Learn how to process Sentinel-1 SAR data for flood detection. - Understand how SAR can penetrate clouds and detect water bodies. - Apply GIS techniques to create flood hazard maps. - Gain experience in disaster risk assessment and response planning.

Required Tools

Sentinel-2 data

QGIS

You can download required data from the link below

You can download the application from the link below

Steps to Complete Exercise 2

Step 1

Step 3

Step 2

Step 4

Step 5

Step 6

03

Exercise 3: Landslide Risk Assessment Using DEM & Sentinel-2

03

Objective

This exercise trains students in landslide risk assessment by integrating Digital Elevation Models (DEM) and Sentinel-2 vegetation data (NDVI). By analyzing terrain slope, aspect, and vegetation cover, students will identify landslide-prone areas and create a hazard risk map. This approach supports the PREVENT project’s goal of using deep technology for natural disaster prevention by developing analytical skills for terrain risk modeling and hazard prediction.

Expected Outcomes

- Learn how to assess landslide risk using terrain and vegetation data. - Gain skills in DEM processing and NDVI analysis. - Understand how terrain slope, aspect, and vegetation loss contribute to landslides. - Develop GIS-based models for landslide prediction and risk assessment.

Required Tools

Sentinel-2 data

QGIS

You can download required data from the link below

You can download the application from the link below

Steps to Complete Exercise 3

Step 1

Step 3

Step 2

Step 4

Step 5

Step 6

04

Optimal Placement of Forest Observation Towers Using GIS

04

Objective

This exercise focuses on spatial analysis for wildfire prevention by identifying optimal locations for forest observation towers. Using DEM, land cover data (CORINE), and road networks (OpenStreetMap), students will determine suitable sites that maximize visibility and accessibility. This exercise supports the PREVENT project’s goal of leveraging geospatial analysis for disaster prevention and early warning systems.

Expected Outcomes

- Learn how to apply spatial analysis to wildfire prevention. - Gain skills in GIS-based site selection. - Understand buffering techniques for accessibility and resource management. - Develop visibility analysis models for fire observation planning.

Required Tools

QGIS

CORINE

OpenStreetMap

Sentinel-2 data

You can download the application from the link below

Land cover data to extract forested areas.

Layers for roads and water bodies.

You can download required data from the link below

Steps to Complete Exercise 3

Step 1

Step 3

Step 2

Step 4

Step 5

Step 6

Course completed!

SAR Data Processing

Flooded areas have low reflectivity (dark values) in SAR data. The difference between before and after flooding can reveal flooded areas. Procedure: 1. Open the Raster Calculator. 2. Calculate the difference between the two pictures: Copy code "After@1" - "Before@1" 3. Save the result as Flood_Difference.tif.

Visibility analysis

1. Go to Raster > Terrain Analysis > Viewshed. 2. Define the coordinates of possible observatory positions from the result. 3. Save as Visibility_Analysis.tif.

Cartographic Overlay

Combine the above criteria: • Open the Raster Calculator: Copy code ("High_Elevation@1" = 1) AND ("Low_Slope@1" = 1) AND ("Road_Buffer@1" = 1) AND ("Forest_Area@1" = 1) • Save as Fire_Tower_Locations.tif.

Introduction to QGIS

1. Open QGIS. 2. Go to Layer > Add Raster Layer. 3. Enter the Sentinel-1 data: o Before (before the flood): the relevant file. o After (after the flood): the corresponding file4. Make sure they are displayed correctly.

Calculating the area

1. Go to Raster > Zonal Statistics. 2. Calculate the area of risk areas (in hectares).

NDVI calculation

NDVI is a vegetation index calculated as: NDVI=(NIR-RED)(NIR+RED)NDVI=(NIR+RED)(NIR-RED) Procedure: 1. Open the Raster Calculator: o Enter the equation: Copy code ("B08@1" - "B04@1") / ("B08@1" + "B04@1") o Save as NDVI.tif:2. Use thresholds (e.g., NDVI < 0.3) to identify areas with low vegetation.

Data collection

1. DEM (Digital Terrain Model): o Go to the Copernicus DEM Hub. o Download the DEM data for the area of interest (e.g., Pindos). 2. CORINE Land Use: o Visit the Copernicus Land Monitoring Service. o Download the CORINE data for your region. 3. Road and Hydrographic Network: o Download data from OpenStreetMap: ▪ For roads: Use the QuickOSM tool in QGIS. ▪ For hydrographic network: Repeat the same procedure.

Create a Risk Map

1. Go to Project > New Print Layout. 2. Create a map: o Show landslide risk areas. o Add memo, title and scale. 3. Export the map as a PDF.

Identifying Flooded Areas

Thresholding can be used to identify flooded areas by the difference in reflectivity. Procedure: 1. Open the Raster Calculator again. 2. Enter the equation: Copy code ("Flood_Difference@1" < -5) * 1 + ("Flood_Difference@1" >= -5) * 0 o Price 1: Flooded areas. o Value 0: Non-flooded areas. 3. Save the result as Flooded_Areas.tif.

Download data

1. Visit the Copernicus Data Space Ecosystem . 2. Create an account (if you don't already have one). 3. Search for Sentinel-2 data for the Marathon area: o Area Coordinates: ▪ North Latitude: 38.15°N ▪ South Latitude: 38.05°N ▪ East Longitude: 24.05°E ▪ Western Longitude: 23.95°E o Dates: ▪ Before the fire: 1 July 2024 ▪ After the fire: 20 August 2024 o Here is the Sentinel-2 Level-2A product. o Filter for <10% cloud coverage. 4. Download bands B04 (red) and B08 (near infrared).

here

Identifying Burnt Areas

1. Apply thresholding for burnt areas: o Open the Raster Calculator. o Enter the following equation: Copy code ("NDVI_Change@1" < -0.3) * 1 + ("NDVI_Change@1" >= -0.3) * 0 o Save as Burned_Areas.tif. 2. Load the new raster onto the map. o Burnt areas have a value of 1.

Create Siting Criteria

Altitude • Go to Raster > Terrain Analysis > Slope and calculate the slope. • Then open the Raster Calculator: Copy code "DEM@1" > 500• Save as High_Elevation.tif. Ground slope • From the Slope output, apply a filter: Copy code "Slope@1" < 10• Save as Low_Slope.tif. Buffer Belts 1. Go to Vector > Geoprocessing Tools > Buffer. o Set a zone of 500 meters from roads (Road_Buffer). o Define a zone 300 meters from the hydrographic network (Water_Buffer).

Download data

1. Visit the Copernicus Data Space Ecosystem . 2. Create an account (if you don't already have one). 3.Set the area of interest (AOI): o Select an area affected by floods, e.g. Evros region. o Coordinates: ▪ North Latitude: 41.25°N ▪ South Latitude: 40.85°N ▪ East Longitude: 26.85°E ▪ Western Longitude: 26.40°E 4. Set the time interval: o Before the flood: 1 September 2024. o After the flood: 10 September 2024. 5. Filter the data: o Select Sentinel-1 GRD (Ground Range Detected) data. o Polarity: VV or VH. 6. Download the files.

here

Create a map

1. Go to Project > New Print Layout. 2. Create a map that includes: o Flooded areas (colour blue). o Legend, title ("Map of Flooded Areas of Evros"), scale and data source. 3. Export the map as a PDF or image.

Calculating the area of burnt areas

1. Go to Raster > Zonal Statistics. 2. Select the Burned_Areas.tif file. 3. Calculate the total area of the burnt areas.

Compare NDVI (Analysis of Variance)

1. Open the Raster Calculator again. 2. Calculate the NDVI difference (after - before): o Enter the equation: Copy code "NDVI_After@1" - "NDVI_Before@1" o Save the result as NDVI_Change.tif. 3. See the new raster: o Low or negative values indicate loss of vegetation (possible burnt areas).

Download and Import the Data

1. DEM (Digital Terrain Model): o Visit the Copernicus DEM Hub. o Download data for the region of interest, e.g., Pindos region. 2. Sentinel-2: o Visit the Copernicus Data Space Ecosystem. o Download the B04 (red) and B08 (NIR) bands for the NDVI calculation. o Set time interval: before and after heavy rainfall. 3. Precipitation maps: o Download precipitation data (e.g., WorldClim) for the same region. 4. Importing Data into QGIS: o Go to Layer > Add Raster Layer. o Enter DEM and Sentinel-2 data.

Correlation of Data

By combining slope, orientation and NDVI, we can create a landslide risk map. Procedure: 1. Open the Raster Calculator. 2. Enter the equation for landslide risk: Copy code ("Slope@1" > 30) * 1 + ("NDVI@1" < 0.3) * 1 3. Save as Landslide_Risk.tif.

Import data into QGIS

1. Open QGIS. 2. Go to Layer > Add Raster Layer. 3. Load bands B04 and B08: o Before (before the fire): enter bands B04 and B08 for the date July 1, 2024.o After (after the fire): enter bands B04 and B08 for the date August 20, 2024.4. Make sure they are displayed correctly on the map.

Calculating the area of flooded areas

1. Go to Raster > Zonal Statistics. 2. Select the Flooded_Areas.tif file. 3. Calculate the total area of areas with a value of 1.

Create a map

1. Go to Project > New Print Layout. 2. Create a thematic map: o Display the suggested locations. o Add memo, title, and scale. 3. Export the map as a PDF or image.

NDVI calculation

1. Open the Raster Calculator: o Go to Raster > Raster Calculator. 2. Calculate the NDVI for the image before the fire: o Enter the equation: Copy code ("B08@1" - "B04@1") / ("B08@1" + "B04@1") o Save the result as NDVI_Before.tif. 3. Calculate the NDVI for the image after the fire: o Enter the same equation. o Save the result as NDVI_After.tif.

Slope and Orientation Analysis

Slope measures the steep elevation of the ground, while Aspect shows the direction in which the ground is facing. Procedure: 1. Go to Raster > Terrain Analysis > Slope: o Select the DEM file. o Save as Slope.tif. 2. Go to Raster > Terrain Analysis > Aspect: o Select the DEM file. o Save as Aspect.tif.

Preparing Data in QGIS

1. Open QGIS and load all the files: o Go to Layer > Add Layer > Add Raster/Vector Layer. o Enter the DEM, CORINE, and OpenStreetMap data. 2. Land Use Processing: o Filter the forest areas (CORINE): ▪ Go to Vector > Geoprocessing Tools > Vector > Geoprocessing Tools > Extract by Attribute. ▪ Apply filter: Classes 311, 312, 313.

Visualization & Map Creation

1. Open Project > New Print Layout. 2. Add the thematic map: o Show the burnt areas. o Add a footnote, title ("Burnt Areas of Marathon"), scale and data source. 3. Export the map as a PDF or image.